issue: June 2003 APPLIANCE Magazine Computer Simulation Computer Simulation Helps Meet Oven Specs While Reducing Cost	Share  Printable format  Email this Article  Search
The use of computer simulation helped to demonstrate how a new oven design could meet the UnderwriterÕs Laboratory (UL) outside surface temperature requirements with an insulation material substantially less costly than the main alternative.

Testing of early prototypes showed that the door outside surface temperature needed to be reduced. Engineers considered several options and were focusing on the use of a micro-porous insulation material that would have cost approximately U.S. $30 per unit more than the fiberglass that is normally used. At this point, computational fluid dynamics (CFD) was used to evaluate a wide range of design alternatives and generate surface temperature predictions, the accuracy of which was later verified with experimental measurements. The simulations showed that switching from fiberglass to the newer material actually had very little impact on outer surface temperature. On the other hand, engineers learned that they could reduce the temperature to acceptable limits by increasing the width of the air-wash, the gap separating the inner and outer sides of the oven door, by a few millimeters. This minor change did not increase the cost of the product. The ability to quickly and inexpensively check the performance of many different designs without having to build and test a prototype for every design variation was the key to the success of the project.

Figure 1 shows a typical oven door construction. The inner part of the door consists of a liner, a window-pack, insulation, and a baffle to retain the insulation. The window-pack consists of two parallel glass pieces (G1 and G2) that, along with metal brackets, enclose an air cavity. The outer side of the door consists either entirely of glass or glass in combination with plastic or metal trim. The handle used to open the door is not shown in this figure. The inner and outer sides of the door are separated by a distance often called the air-wash, a space that uses natural convection for cooling. Air enters at the bottom of the air-wash, rises as it is heated by the hot walls on both sides, and exits through trim vents near the top of the outer door.

Complicated Physical Phenomena

Heat transfer from the oven to the door involves the interaction of a number of complicated physical phenomena. The inner door liner is heated from the oven, obtaining its maximum temperature during the cleaning cycle. Heat is then transferred by conduction from the liner through the insulation to the baffle. If there is no insulation between the liner and baffle, heat will be transferred through radiation and, if there is air movement, convection as well. In the window-pack region, heat is transferred from the oven side to the outer glass of the pack through radiation and conduction. A key design goal is to optimize the width of the window-pack so that air circulation is either eliminated or reduced, thereby minimizing the convective heat transfer inside the pack. Radiation is the dominant mode of heat transfer from inner door to outer door across the air-wash, with cooling accomplished by buoyancy-driven natural convection. The outer surface of the outer door is cooled by natural convection and radiation to the environment of the kitchen.

The complexity of the physics and geometries involved makes it impossible in most practical designs to predict performance using conventional engineering calculations. As a result, the traditional approach to oven design involves creating an initial concept design based on engineering judgment and experience, building and testing a prototype, and then, based on the results, reconfiguring the design and repeating the entire process again and again until an acceptable design is achieved. The problems with this approach include the relatively high cost and amount of time required to build and test prototypes. Another problem is that cost and time constraints typically make it impossible to optimize a design. Instead, engineers typically face budget constraints and product introduction targets, and often must settle for the first design they hit upon that is good enough to meet the product specifications.

Move to Computer Simulation

Maytag Cleveland Cooking Products in Cleveland, TN, U.S. has replaced this process with computer simulation because it makes it possible to evaluate a wide range of design alternatives in a relatively short period of time. CFD provides fluid velocity, pressure, and temperature values throughout the solution domain with complex geometries and boundary conditions. As a part of the analysis, a designer may change the geometry of the system or the boundary conditions, such as inlet pressure or temperature, and view the effect on the fluid flow and thermal characteristics. This makes it possible for engineers to investigate the effects of changing certain parameters, such as the insulation compartment geometry, the thickness of the insulation, the type of insulation, and the size and shape of entry and exhaust vents.

The commercially available CFD software, FLUENT (Lebanon, NH, U.S.), was used to obtain results discussed here. FLUENT is a provider of computational fluid dynamics (CFD) software and consulting services. FLUENT’s software is used for simulation, visualization, and analysis of fluid flow, heat and mass transfer, and chemical reactions.

It is important to note that CFD does not replace experimental methods; rather it provides a method of determining optimum values for critical design parameters prior to experimental validation. Consequently, it reduces the number of design iterations and physical building of test models (prototypes) that would otherwise be built.

The CFD analysis of the initial design for the oven provided the engineering team with a far better understanding of the sensitivity of the design to the key parameters than could have been gained by physical testing alone. Engineers began by simulating the prototype exactly as it was built and tested, and compared FLUENT predictions to the physical testing results. As Figure 2 shows, the measured temperatures closely matched the CFD simulation. The next step was to evaluate various design changes in an effort to solve the immediate problem of reducing the outside surface temperature of the door, while avoiding unnecessary cost increases.

Evaluating the Effect of Design Parameters

To check the efficacy of using insulation between the liner and baffle at the top of the door, engineers changed the CFD model assumptions so that the movement of air in region B was either allowed or not allowed in the CFD calculation. Figure 3 shows the temperature profile for the case where air was not free to move. The results showed that there is virtually no difference in outer door temperature under the two different assumptions. Given that still air is the best insulation—with the exception of a vacuum—engineers concluded that filling region B with insulation wouldn’t reduce the outer door temperature.

Next, the CFD model was changed to predict the temperature profile for the outer door with a 1/8-in fiberglass insulation inserted behind the outer door metal trim in the air-wash region. Contrary to intuition, the temperature of the outer door rises by about 8°C due to this trim insulation (see Figure. 4). The reason for this rise in temperature is that while the trim insulation blocks the conduction heat from coming into the trim, it also restricts the flow in the air-wash. The net effect is an undesirable rise in the outer door temperature.

Determining that this potential solution was not on the right track during the early stages of the design process saved time and money because the engineers’ attention could be redirected to other alternatives.

Comparing Different Types of Insulation

The next step was comparing the performance of two insulation types, fiberglass and micro-porous. Fiberglass is the insulation material most commonly used in ovens and costs about $2 to $5 per oven. Micro-porous insulation is a new product that offers some advantages, but the cost could range from $24 to $38 per oven, depending upon the size needed. The CFD analysis showed that in this specific application, the improved insulating characteristics of micro-porous insulation yielded only a small improvement of 1°C to 2°C in the outer door trim temperatures as shown in Figure 5. This improvement was not enough to meet UL specifications. Based on the analysis, engineers decided to use the fiberglass insulation, thus avoiding an additional cost of approximately $30 per unit.

Figure 5. Outer door temperature profiles due to fiberglass and micro-porous insulation. (CLICK to see enlarged image.)

Figure 6 shows the temperature profiles for a set of high emissivity values (curve 1) and for a set of low emissivity values (curve 2). The measured temperature points lie in between these two curves, once again showing the validity of the CFD model. The temperature levels with the set of low emissivity numbers (curve 2) exceeded the level allowed by UL.

Next, engineers tried increasing the air-wash width by 6.35 mm (1/4 in) to assess the effect on the outer door temperature. Figure 6 (curve 3) shows that there is a drop of about 21°C in the outer door temperature in the trim located in the upper region of the outer door. This portion of the analysis demonstrated that a small increase in air-wash width, which has virtually no associated cost, provides a far greater reduction in outside surface temperature than a move to a more expensive insulating material. The results also demonstrate how computer simulation can improve appliance design by allowing engineers to quickly determine the impact of various design parameters.

Figure 6. Effect of air-wash width on outer door temperature. (CLICK to see enlarged image.)

Computer simulation is an idea that’s time has come for appliance design. New modeling techniques provide engineers with the ability to model the performance of design concepts with reasonable accuracy without having to build a prototype. This makes it possible to evaluate many more designs, usually resulting in a substantial improvement in performance. At the same time, the lower cost and shorter lead times of simulation provide faster time to market and reduced development costs. In addition, since CFD simulation provides even more design data than physical testing, it is an indispensable option for engineers involved in appliance design.

This information is provided by Prabhat Tekriwal, Maytag Cleveland Cooking Products, Cleveland, TN, U.S.